监测接收机中调制识别的应用研究与系统实现
发布时间：2018-08-22 12:36
【摘要】：伴随着无线电技术的快速发展,为了更有效率地利用和管理无线电频谱资源,无线电频谱监测系统越来越多的被应用到监测任务之中。监测系统面临的电磁环境越来越复杂,因此对监测系统的功能需求也在不断增加。在众多的功能需求中,对信号调制类型的自动识别逐渐发展为监测系统必备的功能之一。本文在监测接收机系统硬件平台上对调制识别功能进行了深入的研究,针对调制识别功能在系统DSP模块中的实现,本文主要进行了以下四部分的研究工作。首先针对监测接收机系统进行了设计,其中包括DSP与PC和FPGA的通信机制,DSP中多任务时线程之间的调度,以及调制识别的工作模式。整个DSP系统内部的线程本文采用了DSP/BIOS操作系统对其进行调度,文中详细阐述了各个工作模式下线程的调度情况。整个接收机系统的通信接口分别是利用McBSP和GPIO与FPGA通信,利用TCP/IP协议与PC进行网络通信。本文对调制识别算法进行了分析,采用了比较利于实现的数字混频正交变换对信号的瞬时信息进行提取,再利用判决树对信号的类型进行分类。在matlab中,利用接收机采集到的原始数据进行了仿真实验,分析了7种信号的特征参数变化趋势,并给出了可行的判决树方案。最后讨论了在系统实现时,针对该算法进行的适当调整并结合各种应用情况进行分析。接下来本文对整个系统运行时,可能会出现的系统不稳定因素进行了分析,包括器件本身存在的问题,和程序可能出现的异常情况。利用系统现有的资源对异常情况的处理进行了研究,给出了系统自检和异常处理的实现方案。最后,基于整个监测接收机系统进行了集成联调。其中包括对工作模式的测试,以及调制识别算法在系统中针对12种信号的识别率测试。测试结果表明调制识别算法在DSP中可以正常工作,同时能够很好地对信号进行识别,在信噪比不小于10dB时系统的识别率在91%以上。
[Abstract]:With the rapid development of radio technology, in order to use and manage radio spectrum resources more and more efficiently, radio spectrum monitoring system is more and more used in monitoring tasks. The electromagnetic environment of the monitoring system is becoming more and more complex, so the functional requirements of the monitoring system are also increasing. Among the many functional requirements, automatic recognition of signal modulation types has gradually developed into one of the essential functions of the monitoring system. In this paper, the modulation recognition function is deeply studied on the hardware platform of the monitoring receiver system. Aiming at the realization of the modulation recognition function in the DSP module of the system, the following four parts of research work are mainly carried out in this paper. First of all, the monitoring receiver system is designed, which includes the scheduling of multi-task threads in DSP, PC and FPGA communication mechanism, and the mode of modulation identification. The threads in the whole DSP system are scheduled by the DSP/BIOS operating system in this paper, and the scheduling of the threads in each working mode is described in detail. The communication interface of the whole receiver system is to use McBSP and GPIO to communicate with FPGA, and to communicate with PC by TCP/IP protocol. In this paper, the modulation recognition algorithm is analyzed, the digital mixing orthogonal transform is used to extract the instantaneous information of the signal, and then the type of the signal is classified by the decision tree. In matlab, the simulation experiment is carried out with the raw data collected by the receiver, and the change trend of the characteristic parameters of seven signals is analyzed, and a feasible decision tree scheme is given. Finally, the proper adjustment of the algorithm and its application are discussed. Then this paper analyzes the system instability factors which may occur when the whole system is running, including the problems existing in the device itself and the abnormal situation that may occur in the program. Based on the existing resources of the system, the processing of abnormal cases is studied, and the realization scheme of system self-checking and exception handling is given. Finally, the integrated tuning is carried out based on the whole monitoring receiver system. It includes the test of working mode and the recognition rate test of modulation recognition algorithm for 12 kinds of signals in the system. The test results show that the modulation recognition algorithm can work normally in DSP and can recognize the signal well. The recognition rate of the system is more than 91% when the SNR is not less than 10dB.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN98
本文编号：2197122
[Abstract]:With the rapid development of radio technology, in order to use and manage radio spectrum resources more and more efficiently, radio spectrum monitoring system is more and more used in monitoring tasks. The electromagnetic environment of the monitoring system is becoming more and more complex, so the functional requirements of the monitoring system are also increasing. Among the many functional requirements, automatic recognition of signal modulation types has gradually developed into one of the essential functions of the monitoring system. In this paper, the modulation recognition function is deeply studied on the hardware platform of the monitoring receiver system. Aiming at the realization of the modulation recognition function in the DSP module of the system, the following four parts of research work are mainly carried out in this paper. First of all, the monitoring receiver system is designed, which includes the scheduling of multi-task threads in DSP, PC and FPGA communication mechanism, and the mode of modulation identification. The threads in the whole DSP system are scheduled by the DSP/BIOS operating system in this paper, and the scheduling of the threads in each working mode is described in detail. The communication interface of the whole receiver system is to use McBSP and GPIO to communicate with FPGA, and to communicate with PC by TCP/IP protocol. In this paper, the modulation recognition algorithm is analyzed, the digital mixing orthogonal transform is used to extract the instantaneous information of the signal, and then the type of the signal is classified by the decision tree. In matlab, the simulation experiment is carried out with the raw data collected by the receiver, and the change trend of the characteristic parameters of seven signals is analyzed, and a feasible decision tree scheme is given. Finally, the proper adjustment of the algorithm and its application are discussed. Then this paper analyzes the system instability factors which may occur when the whole system is running, including the problems existing in the device itself and the abnormal situation that may occur in the program. Based on the existing resources of the system, the processing of abnormal cases is studied, and the realization scheme of system self-checking and exception handling is given. Finally, the integrated tuning is carried out based on the whole monitoring receiver system. It includes the test of working mode and the recognition rate test of modulation recognition algorithm for 12 kinds of signals in the system. The test results show that the modulation recognition algorithm can work normally in DSP and can recognize the signal well. The recognition rate of the system is more than 91% when the SNR is not less than 10dB.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN98
【参考文献】
中国博士学位论文全文数据库 前1条
1 徐毅琼;数字通信信号自动调制识别技术研究[D];解放军信息工程大学;2011年
中国硕士学位论文全文数据库 前1条
1 徐婷;网络化数字监测接收机信号处理算法研究及DSP实现[D];电子科技大学;2012年
，本文编号：2197122
本文链接：https://www.wllwen.com/kejilunwen/wltx/2197122.html
